• Options
• Login
• Desktop Style

Matching Plant Images Internally

• Index

  • The Voynich Ninja
  • Voynich Research
  • Imagery
  • Matching Plant Images Internally
• Matching Plant Images Internally

• RE: Matching Plant Images Internally

  Aga Tentakulus > 23-09-2023, 11:17 AM

  Now I may remember that I wrote the following at Bax about the root pages.
  "If I want to assign a root to a plant, I also need the leaf."
  That is exactly what the VM author did.
  So the leaf must also match the plant and not just the root if I want to be more sure.

• RE: Matching Plant Images Internally

  bi3mw > 23-09-2023, 01:30 PM

  (23-09-2023, 10:45 AM)Koen G Wrote: You are not allowed to view links. Register or Login to view.Yeah that's a good example of a clear match and a clear mismatch. My intuition would be to work with a smaller subset of the most clear couples.

  
  I agree with this. Also, it would be useful to write the folio number next to the examples.

• RE: Matching Plant Images Internally

  Mark Knowles > 23-09-2023, 02:12 PM

  (23-09-2023, 09:55 AM)Aga Tentakulus Wrote: You are not allowed to view links. Register or Login to view.I would describe this example as not comparable.
  On one side there is a root with three fingers, on the other side there are three roots of plants forming a group. I do not see any correspondence here.

  
  It is certainly understandable why someone would think these two drawings are not a match. And indeed they may well not be a match.
  
  So why have I matched them? Because starting with the drawing of the root I looked through all the plant drawings to find the one that I thought most similar or most reminiscent and I felt that the herbal drawing stood out. I think one's attitude about the type of artistic representation that the author is employing is important; whether scientific/botanical accuracy or symbolic representation. I match these two, because from the comparison of the illustrations I think it quite probable or possible, though clearly not certain, that the artist was trying to represent the same plant. I certainly prefer to include this possible match in my list than to ignore it, that is why I chose to include it. There are other potential matches that I chose not include as I felt I could not justify to myself so doing.

• RE: Matching Plant Images Internally

  Mark Knowles > 23-09-2023, 03:59 PM

  (23-09-2023, 01:30 PM)bi3mw Wrote: You are not allowed to view links. Register or Login to view.

  (23-09-2023, 10:45 AM)Koen G Wrote: You are not allowed to view links. Register or Login to view.Yeah that's a good example of a clear match and a clear mismatch. My intuition would be to work with a smaller subset of the most clear couples.

  
  I agree with this. Also, it would be useful to write the folio number next to the examples.

  
  I have started to organise this data, but I have some way to go and other tasks to accomplish, so I thought I would make my work public and possibly someone else might decide to organise this into a spreadsheet.
  
  Not only the folio number, but also the row&column of the botanical drawing within the botanical folio would be useful.

• RE: Matching Plant Images Internally

  Mark Knowles > 23-09-2023, 04:28 PM

  Being able to match up plant drawings may give insights into the order in which plants appear in different parts of the manuscript. Is the ordering consistent? Can botanical plant ordering give insights into herbal folio ordering?
  
  But mostly textual comparison of individual plant labels to matching plant labels to herbal text seems the most worthy of careful scrutiny. One would hope to find some kind of connection/similarity/relationship between some labels and other text.

• RE: Matching Plant Images Internally

  Mark Knowles > 23-09-2023, 05:08 PM

  Someone might want to give a score to each match indicating how close the match is. It might also help to have a note against each match explaining or justifying why that match was chosen.
  
  Once the information is in a spreadsheet it would be possible to perform an inner join to connect the matches for the same plant.

• RE: Matching Plant Images Internally

  merrimacga > 23-09-2023, 09:33 PM

  (23-09-2023, 05:08 PM)Mark Knowles Wrote: You are not allowed to view links. Register or Login to view.Someone might want to give a score to each match indicating how close the match is. It might also help to have a note against each match explaining or justifying why that match was chosen.

  
  Perhaps it would be helpful to poll members about each match you have identified. I looked into this and found a You are not allowed to view links. Register or Login to view. but you needed Koen's help creating that one. If it is still not possible for members, rather than moderators, to create a poll, then please disregard. But if members can start a poll, it appears as though it can only be done when creating a thread, rather than in later posts to that thread, and only one poll question allowed per poll and only one poll per thread, though that question can have multiple response options (not sure if there is a limit). In this case, you would need a separate poll thread to compare each match of 2 images you found. Where you have 3 matching images (or more), you will have a risk of lower agreement response or you will need to break out those matches so only 2 images are compared per poll (i.e. image 1 vs image 2, then image 1 vs image 3, then image 2 vs image 3). The thread itself could link back to this one (and vice versa) and include the images with folio number labels for comparison plus the justification note you mentioned. I recommend not setting a poll timeout but your thread verbiage should give an ample deadline or a minimum volume of responses and there is a poll edit option with a checkbox to close the poll manually at any time. I would also suggest setting the responses so users can indicate the percentage or degree to which the images match (e.g. no match, partial match, close match and full match OR 0%, 25%, 50%, 75% and 100% match OR match strongly agree, agree, neither agree nor disagree, disagree and strongly disagree). Respondents can then add comments as new posts to the poll thread, keeping this thread clear for responses not related to the polls. I have posted the MyBB help link at bottom for more details on creating polls.
  
  Admittedly, all these poll threads will be cumbersome but you could, and it would be advisable, to spread them out so that you don't have more than a handful of active ones running at one time. It's also a good idea, assuming it is possible for members to create polls, to run the first poll also as a test, just to be sure this will work well moving forward for the rest of the comparisons. Then if you need help setting up and managing the polls, I would be happy to help so you don't have to do all the work and so you can concentrate on finishing the spreadsheet.
  
  Just a thought.
  
  MyBB Documentation: Polls
  You are not allowed to view links. Register or Login to view.
  
  (You may also right click on your previous poll page and click on the pop up menu option to view source coding but I don't know if this site will allow you to HTML code the poll from scratch.)

• RE: Matching Plant Images Internally

  Koen G > 23-09-2023, 11:01 PM

  If we don't want to bother with polls, we could also set up a shared google drive sheet. Mark kind of numbered his plants, so those could be used as reference. Then each member can fill in a column. We would just need to agree on a scoring system. Maybe something like:
  
  3: best kind of match you will find in the manuscript
  2: it has issues, but I would still include it
  1: I see some similarity but would probably not include it
  0: total mismatch, it would be a bad idea to include this
  
  Then we can average the scores and afterwards determine a cutoff point.
  
  If there is agreement, we can use this document:
  
  You are not allowed to view links. Register or Login to view.
  
  Note that for now this is only for sifting through the proposed matches. We can add folio info etc later.
  
  (It is late here, I will check back here and enter my scores tomorrow. Feel free to fill in the doc meanwhile).

• RE: Matching Plant Images Internally

  Koen G > 24-09-2023, 12:33 PM

  I filled in my column. Sometimes it's quite subjective though, so if we want to get a decent selection we need a handful of people to do their own.
  
  Edit: I changed the permissions, normally everybody should be able to edit the file now.

• RE: Matching Plant Images Internally

  bi3mw > 24-09-2023, 12:52 PM

  Hi Koen, what is match no. 4 ? I understand that it can only go from 0 to 3.
• Next Oldest Next Newest